Sample characteristics: Uganda

Census characteristics
	1991	2002
Title of census	1991 Population and Housing Census	2002 Uganda Population and Housing Census (with agricultural module)
Census agency	Uganda Bureau of Statistics	Uganda Bureau of Statistics
Population universe	All persons who are in Uganda the night of the census, regardless of their nationality.	All persons who are in Uganda the night of the census, regardless of their nationality.
De jure or de facto	De facto	De facto (all) and de jure (heads only)
Enumeration unit	Households	Households
Census day	January 11, 1991	September 12, 2002
Field work period	7 days, with no curfew	7 days, with a curfew in the capital on the first day
Enumeration forms used	Schedule A: short form and Schedule B: long form	Unique census questionnaire (no specific name)
Type of field work	Face to face interview	Face to face interview
Coverage	—	—

Back to Top

Microdata sample characteristics
	1991	2002
Microdata source	—	—
Sample design	A sample of approximately 10% of the rural Enumeration areas where a long questionnaire was administered to the households (excluding INSTITUTIONS) these areas (LONG_RURAL), while ALL urban areas (11.3% of the population) were enumerated with a long questionnaire (URBAN). Thus the data set consists of these two sets (LONG RURAL and URBAN). The data also includes a variable called weight which is record-specific. This is the weight which you should use, and the information you will get will be representative of the Household Population ONLY (= 98%).	A 10% systematic sample of questionnaire records (households and institutions) was taken. These could have been households or institutions. A uniform weight of 10 should thus be attached to each record and the resultant population will 99.98% of the non-sample population.
Sample unit	Enumeration Area	Questionnaire record
Sample fraction	10%	10%
Sample size (person records)	1,548,460	2,497,449
Sample weights	Weights computed by national statistical office taking into consideration five key variables. Urban areas are sampled at 10 times the rate of rural areas.	Self-weighting (expansion factor=10)

Back to Top

Units Identified in Microdata
	1991	2002
Dwellings	Yes	Yes
Vacant units	No	No
Households	Yes	Yes
Individuals	Yes	Yes
Group quarters	Yes	Yes
Settled/Unsettled Population	Yes	Yes
Special populations	Yes - visitors	Yes - homeless
Smallest geography	County	County

Back to Top

Unit Definitions (Census)
	1991	2002
Dwellings	The accommodation occupied by one household is the dwelling unit.	A dwelling unit is the unit actually occupied by the household.
Households	A household is a group of persons who normally live and eat together, regardless of whether they are related.	A household is defined as a group of persons who normally eat and live together. If a man has two or more wives and they and their children live and eat together, they form one household.
Group quarters	Sometimes groups of people live together but cannot be said to belong to a household. Persons in hospitals, colleges, barracks and prisons are examples.	Sometimes groups of people live together but cannot be said to belong to a household. Persons in hospitals, colleges, barracks and prisons are examples.
Unsettled population	Floating population refers to those who will not spend census night in households, institutions or hotels. They include persons who are travelling on census night, those in transit at airports or on ships or in railway stations. They include also beggars, vagrants and other homeless people who spend the night at bus parks, on the streets or similar places.	Floating population refers to persons who will not spend the census night in any Household, institutions or hotels. They include persons who were traveling on the census night, those in transit at airports or on ships in landing sites or in railway stations. Homeless are people who do not have any form of formal shelter over their heads e.g. those sleeping on verandahs, condemned, abandoned or partially demolished structures. They also include beggars, vagrants, street children and people who spend the night at bus parks, on the streets or similar places.

Back to Top